© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2823
    Prev
    Next
    INDEX
    Explanations

    [MASK] or placeholder

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on tokens conveying numeric data or quantitative measures—e.g. statistics, counts, percentages, or other number‐related terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ,|
    0.61
     desi
    0.54
    च्
    0.54
     estava
    0.53
    નુ
    0.52
     aveva
    0.51
    یکیشن
    0.51
    에요
    0.49
    ятся
    0.49
     lesz
    0.49
    POSITIVE LOGITS
    __.
    0.66
     ביותר
    0.65
    ‍♂️
    0.63
    ‍♀️
    0.63
    __,
    0.62
    __:
    0.62
     Dhabi
    0.60
    __);
    0.60
     большинстве
    0.60
     plupart
    0.59
    Activations Density 0.375%

    No Known Activations